Automatic detection of pneumonia in chest X-ray images using textural features

• Published in: Computers in biology and medicine 2022-06, Vol.145, p.105466-105466, Article 105466
• Main Authors: Ortiz-Toro, César, García-Pedrero, Angel, Lillo-Saavedra, Mario, Gonzalo-Martín, Consuelo
• Format: Article
• Language: English
• Subjects: Algorithms; Artificial Intelligence; Biomarkers; Chest; Chest imaging; Child; Coronaviruses; COVID-19; COVID-19 - diagnostic imaging; Datasets; Deep Learning; Diagnostic imaging; Fractal dimension; Fractal geometry; Fractals; Histon; Humans; Machine learning; Medical diagnosis; Medical imaging; Pandemics; Pediatrics; Pneumonia; Radiomics; SARS-CoV-2; Sensitivity; Superpixels; Support vector machines; X-ray; X-Rays
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/j.compbiomed.2022.105466

Fast and accurate diagnosis is critical for the triage and management of pneumonia, particularly in the current scenario of a COVID-19 pandemic, where this pathology is a major symptom of the infection. With the objective of providing tools for that purpose, this study assesses the potential of three textural image characterisation methods: radiomics, fractal dimension and the recently developed superpixel-based histon, as biomarkers to be used for training Artificial Intelligence (AI) models in order to detect pneumonia in chest X-ray images. Models generated from three different AI algorithms have been studied: K-Nearest Neighbors, Support Vector Machine and Random Forest. Two open-access image datasets were used in this study. In the first one, a dataset composed of paediatric chest X-ray, the best performing generated models achieved an 83.3% accuracy with 89% sensitivity for radiomics, 89.9% accuracy with 93.6% sensitivity for fractal dimension and 91.3% accuracy with 90.5% sensitivity for superpixels based histon. Second, a dataset derived from an image repository developed primarily as a tool for studying COVID-19 was used. For this dataset, the best performing generated models resulted in a 95.3% accuracy with 99.2% sensitivity for radiomics, 99% accuracy with 100% sensitivity for fractal dimension and 99% accuracy with 98.6% sensitivity for superpixel-based histons. The results confirm the validity of the tested methods as reliable and easy-to-implement automatic diagnostic tools for pneumonia. •Three different textural feature extraction techniques, radiomics, fractal dimension and the novel superpixel-based histón are compared for pneumonia detection in thoracic X-ray images.•Classification models generated from two different images datasets, based on different automatic machine learning techniques (KNN, SVM, and RF), are evaluated.•The impact of both hyperparameter optimisation and unbalanced/SMOTE-balanced training sets is also studied.•While all characterisation methods achieve remarkable results, those obtained with both superpixel-based histon and fractal dimension are particularly noteworthy.